Play store
Australian researchers have developed an advanced artificial intelligence (AI) tool capable of predicting the risk of bowel cancer recurrence with greater accuracy, a breakthrough that could help doctors identify high-risk patients earlier and improve treatment outcomes.
The AI algorithm, named SEMIL (Semantically-Enhanced Multiple Instance Learning), analyses routine pathology slides by combining microscopic tissue images with written pathology reports to detect tumour growth patterns associated with cancer relapse, according to a statement released on Wednesday by La Trobe University.
Researchers said the technology has the potential to support more personalised treatment decisions, allowing clinicians to determine which patients require closer monitoring or additional therapies after surgery.
The findings were published in the peer-reviewed journal Gastroenterology.
To develop and validate the tool, researchers analysed more than 1,600 pathology slides and tested the algorithm on 1,220 patients with stage-two bowel cancer across three independent cohorts from several Australian medical institutions.
Bowel cancer, also known as colorectal cancer, is the third most commonly diagnosed cancer worldwide and remains one of the leading causes of cancer-related deaths. Although many patients are successfully treated with surgery, a significant number experience recurrence, making accurate risk assessment crucial.
The study found that the most reliable predictions of cancer relapse were achieved when the AI’s assessment was combined with the expertise of professional pathologists.
Rather than replacing medical specialists, the researchers said the technology is designed to enhance clinical decision-making by providing more consistent and objective evaluations of tumour characteristics.
Lead author Francis Magisson of La Trobe University said the AI system could help identify stage-two bowel cancer patients who are at greater risk of recurrence and would therefore benefit from more intensive follow-up care or chemotherapy.
Under current Australian clinical guidelines, chemotherapy is generally recommended after surgery only for patients considered to have a high risk of their cancer returning.
Supporting precision medicine
Researchers said the new technology represents another step forward in the use of AI to advance precision medicine, where treatment is tailored to the individual characteristics of each patient.
One of the tool’s key advantages is that it can be integrated into existing digital pathology systems without requiring additional laboratory tests, specialised equipment or extra tissue samples.
The researchers believe the approach could improve consistency in pathology reporting, reduce variation in clinical assessments and ultimately help doctors make faster, more informed treatment decisions.
They added that further studies and broader clinical implementation could pave the way for the AI tool to become a routine part of cancer diagnosis and post-surgical care, potentially improving survival rates for thousands of bowel cancer patients worldwide.